In order to achieve a color display, color filters are often used. The most common approach is to add color filters on top of black/white sub-pixels of a pixellated display to display the red, green and blue colors. When a red color is desired, the green and blue sub-pixels are turned to the black state so that the only color displayed is red. When a blue color is desired, the green and red sub-pixels are turned to the black state so that the only color displayed is blue. When a green color is desired, the red and blue sub-pixels are turned to the black state so that the only color displayed is green. When the black state is desired, all three-sub-pixels are turned to the black state. When the white state is desired, the three sub-pixels are turned to red, green and blue, respectively, and as a result, a white state is seen by the viewer.
The biggest disadvantage of using such a technique for a reflective display is that since each of the sub-pixels has a reflectance of about one third of the desired white state, the white state is fairly dim. To compensate this, a fourth sub-pixel may be added which can display only the black and white states, so that the white level is doubled at the expense of the red, green or blue color level (where each sub-pixel is only one fourth of the area of the pixel). Brighter colors can be achieved by adding light from the white pixel, but this is achieved at the expense of color gamut to cause the colors to be very light and unsaturated. A similar result can be achieved by reducing the color saturation of the three sub-pixels. Even with this approach, the white level is normally substantially less than half of that of a black and white display, rendering it an unacceptable choice for display devices, such as e-readers or displays that need well readable black-white brightness and contrast.